search

cBioPortal / cbioportal

cBioPortal for Cancer Genomics
https://cbioportal.org
GNU Affero General Public License v3.0
641 stars 498 forks source link

support single filter parameters for study view #5621

Closed jjgao closed 5 years ago

jjgao commented 5 years ago

http://www.cbioportal.org/beta/newstudy?id=msk_impact_2017#filterAttributeId=CANCER_TYPE&filterValues=Pancreatic Cancer;Breast Cancer

Q: what if there is a ; in the value?

jjgao commented 5 years ago

This is a simple solution to support one filter. Multiple filters will be supported in #4983.

zhx828 commented 5 years ago

@jjgao ; would not allowed to use in the case you mentioned. This is designed for a quick filter, I don't think we should consider too much here. If there is special character, then probably #4983 is a better solution to go.

zhx828 commented 5 years ago

@jjgao for continuous data, this will be the format http://www.cbioportal.org/study?id=nsclc_tcga_broad_2016&filterAttributeId=AGE&filterValues=60-65,70-75

And we only support clinical data for now?

jjgao commented 5 years ago

@zhx828 sounds good.

zhx828 commented 5 years ago

@jjgao I just noticed in the example you posted, filterValues:Pancreatic Cancer;Breast Cancer, different value you want to use semicolon over comma? Also it should be = instead of colon?

I have a working instance here

http://cbioportal-frontend-pr-1935.herokuapp.com/study?id=msk_impact_2017&filterAttributeId=CANCER_TYPE&filterValues=Pancreatic%20Cancer,Breast%20Cancer

http://cbioportal-frontend-pr-1935.herokuapp.com/study?id=lgg_tcga&filterAttributeId=ONCOTREE_CODE&filterValues=OAST

jjgao commented 5 years ago

@zhx828 it should be = instead of : -- just corrected it. Do you have a preference for ; vs ,? Do we have any other url support list and what are they using?

zhx828 commented 5 years ago

@jjgao we use comma for multi-studies. It seems pretty standard way to do. We have similar in patient view navCaseIds

jjgao commented 5 years ago

@zhx828 sounds good!

zhx828 commented 5 years ago

@jjgao should we give some messages when the filterAttributeId is not available in this study?